代码如下：druid版本 1.2.5 数据库sqlserver 分别为group by和order by的解析
String sql1 ="select id, data_base_name, type_name, table_name, operation_name, remark from center..ad_sql_parser where type_name = 1 group by data_base_name";
String sql2 ="select id, data_base_name, type_name, table_name, operation_name, remark from center..ad_sql_parser where type_name = 1 order by data_base_name";
Map<TableStat.Name, TableStat> tableStatMap = getTableNameBySql(sql2);
for (Map.Entry<TableStat.Name, TableStat> tableStatEntry : tableStatMap.entrySet()) {
//1.获取到表名
System.out.println("表名：" + tableStatEntry.getKey().getName());
//2.获取此表的操作名
System.out.println("操作名：" + tableStatEntry.getValue());
}
}

public static Map<TableStat.Name, TableStat> getTableNameBySql(String sql) {
// 数据源类型
DbType dbType = JdbcConstants.SQL_SERVER;
Map<TableStat.Name, TableStat> tables = new HashMap<TableStat.Name, TableStat>();
try {
//格式化 因为sql的关键字，大小写并不敏感，通过这个方法就可以把关键字统一变成大写
String sqlResult = SQLUtils.format(sql, dbType);
logger.debug("格式化后的sql:[{}]", sqlResult);
//解析生产语法树
List stmtList = SQLUtils.parseStatements(sqlResult, dbType);
//判断 解析出来的语法树是否为空
if (CollectionUtils.isEmpty(stmtList)) {
logger.info("语法树为空");
return Collections.emptyMap();
}
//访问 解析语法树
for (SQLStatement sqlStatement : stmtList) {
SQLServerSchemaStatVisitor visitor = new SQLServerSchemaStatVisitor();
sqlStatement.accept(visitor);
//获取到 表名和对应的操作命令
tables = visitor.getTables();
logger.info("druid解析sql的结果集:{}", tables);
}
return tables;
} catch (Exception e) {
logger.error(e.getMessage(), e);
return Collections.emptyMap();
}
}

日志分别打印结果：
格式化后的sql:[SELECT id, data_base_name, type_name, table_name, operation_name, remark FROM center..ad_sql_parser WHERE type_name = 1 GROUP BY data_base_name]
druid解析sql的结果集:{center..ad_sql_parser=Select}
表名：center..ad_sql_parser
操作名：Select

格式化后的sql:[SELECT id, data_base_name, type_name, table_name, operation_name, remark FROM center..ad_sql_parser WHERE type_name = 1 ORDER BY data_base_name]
ERROR com.rails.app.utils.sqlParserUtils - illegal sql expr : center..ad_sql_parser, pos 8, line 1, column 7, token ..